Documentation

Quick Start

Install TraceLLM and capture a first trace.

Quick Start

Get from zero to your first trace in under two minutes. You only need a terminal and Python 3.10+.

Tracey Tip

No database setup, no API keys, no configuration files. Just install and start.

Step 1 — Installation

Install TraceLLM via pip. The package includes the CLI, SDK, and all runtime dependencies.

terminalCopy

bash

pip install tracellm-cli

Tip

Use a virtual environment: python -m venv .venv && source .venv/bin/activate

Step 2 — Start TraceLLM

Start the entire TraceLLM stack with a single command:

terminalCopy

bash

tracellm start

This automatically starts the backend, launches the dashboard, and opens your browser. The SQLite database is auto-created at ~/.tracellm/traces.db on first run.

Note

No configuration needed. SQLite is auto-created. The dashboard opens automatically in your browser.

Step 3 — Run AI Agent

Run a trace against any prompt. TraceLLM captures the full execution — prompts, responses, latency, token usage, and tool calls.

terminalCopy

bash

tracellm trace "Explain transformers"

You will see the trace ID, model response, latency, and token usage printed to the terminal.

Step 4 — View Dashboard

Your browser opened automatically when you ran tracellm start. Navigate to http://localhost:3000 if it did not. Your trace appears in the trace list as soon as it completes.

Dashboard screenshot placeholder

Click on the trace to inspect the full detail: prompt, response, timing breakdown, and every span that was created during execution.